Locally Optimized RANSAC

By Ondřej Chum, Jiri Matas, Josef Kittler. 2003

1,001 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

A Comparative Analysis of RANSAC Techniques Leading to Adaptive Real-Time Random Sample Consensus (2008, methodology): “Some of these strategies [2,3,4] aim to optimize the process of model verification, while others [5,6,7] seek to modify the sampling process in order to preferentially generate more useful hypotheses.”
Image Matching Across Wide Baselines: From Paper to Practice (2020, methodology): “Improvements to this classical framework include local optimization [24], MLESAC [92], PROSAC [23], DEGENSAC [26], GC-RANSAC [12], and MAGSAC [29].”
DSAC — Differentiable RANSAC for Camera Localization (2016, background): “Dozens of variations of RANSAC exist [39, 8, 28, 7].”
A review of multimodal image matching: Methods and applications (2021, background): “Several studies address this issue in a local optimization procedure [314,315], or a progressive growing sampling procedure [311,316,317] to eliminate the need for user-defined thresholds such as inlier–outlier decision.”
Graph-Cut RANSAC (2017, methodology): “In the reported experiments, LO-RANSAC outperforms standard RANSAC in both accuracy and the required number of interations.”
Back to the Feature: Learning Robust Camera Localization from Pixels to Pose (2021, background): “On the other hand, iii) is tackled by classical geometry using feature matching [19, 20, 28] or image alignment [4, 26, 27, 50] and a 3D representation.”
Retrieving landmark and non-landmark images from community photo collections (2010, methodology): “For this, we start from the stored aﬃne model A qp and perform a single step of the “iterative” method of LO-RANSAC.”
PointDSC: Robust Point Cloud Registration using Deep Spatial Consistency (2021, methodology): “RANSAC [24] and its variants [19, 5, 35, 37] are still the most popular outlier rejection methods.”
Total Recall: Automatic Query Expansion with a Generative Feature Model for Object Retrieval (2007, methodology): “For the top 10 hypotheses found, we iteratively use a least-squares re-estimation method on the initially found inliers to generate a full 6 dof affine transformation, returning the best hypothesis as the one with the most inliers after re-estimation [5].”
Energy-Based Geometric Multi-model Fitting (2012, background): “by re-sampling the inliers of the strongest initial models [8].”
Visual Odometry : Part II: Matching, Robustness, Optimization, and Applications (2012, methodology): “In this case, using more points than a minimal set is proved to give better performance (in terms of accuracy and number of inliers) [62], [63].”
Robust monocular SLAM in dynamic environments (2013, background): “Here, we compared our outlier rejection algorithm with the standard RANSAC [13], LoSAC [7], and Multi-GS [5].”
3-D Mapping With an RGB-D Camera (2014, methodology): “To transform a covariance matrix of a point to the coordinate frame of the other sensor pose, we rotate it using R, which is the rotation matrix of the estimated transformation, i.e., Σ i = R T Σ”
MAGSAC: Marginalizing Sample Consensus (2018, methodology): “[3] with an inner RANSAC applied to the inliers.”
Scene Coordinate Regression Forests for Camera Relocalization in RGB-D Images (2013, methodology): “Experimentally we found a pose refinement step [5] to be crucial to achieving accurate localization.”
GMS: Grid-Based Motion Statistics for Fast, Ultra-Robust Feature Correspondence (2017, neutral): “RANSAC [10, 41, 5, 32, 6, 36, 15] can leverage geometric information to alleviate this problem.”
Image Matching from Handcrafted to Deep Features: A Survey (2020, methodology): “The computational cost issue of LO-RANSAC is addressed in Lebeda et al. (2012), where several implementation improvements are suggested.”